stepgrandparents and relational ambiguity

Victor aardvark66 at GMAIL.COM
Thu Jan 22 17:53:23 UTC 2009

I tend to agree that this is the case on underdeterminacy or
underspecification rather than that of ambiguity. In English, "my
sister" is no more ambiguous than "a recycled bottle" (glass or plastic?
green, amber, blue or clear?), but it could be less (I only have one
sister). This can easily be rectified with a modifier. The "in-law"
relationship require a lot more elaboration. FWIW, it's possible to have
two stepmothers, but the ambiguity isn't structural. But two
stepgrandmothers could actually be structurally different. Even in the
case of "cousin" or "parent", as I alluded earlier, there is an
ambiguity, at least, as to gender and it cannot be fixed wit ha simple
modifier. For specification, "parent" needs to be replaced with another
term (mother or father) that is not equivalent to "parent". To specify
"cousin", gender and structural relationship must be narrowed even
before the natural underdeterminacy comes into play. On the other hand,
in some languages, "uncle" and "nephew" words carry a deep structural
ambiguity (essentially creating two classes for most male relatives that
are not parents, children or siblings), in the same sense as equivalent
of "cousin" does in Arabic and, to a lesser extent, in some sub-cultures
in English. In English, "uncle" is ambiguous (degree of kinship, not to
mention the non-relative "uncles"), but, I believe, "nephew" is merely

I mentioned subculture in addition to language. I'll be more specific.
For some speakers, words "uncle", "nephew", "cousin", etc., specify a
precise structural relationship, although they may be ambiguous in other
ways.  In each case, the respective blood relative is "once removed",
i.e., brother of parent, son of sibling, offspring of a parent's
sibling. For others (I suspect, a larger group, for English speakers),
"uncle" might include also a parent's sister's spouse. Yet for others,
"uncle" is any older male relative that is not in direct lineage
(parent, grandparent, etc.) or a male spouse of such a relative,
"cousin" is any non-sibling relative in the same relative age group, and
"nephew" is similar to "uncle", but younger.

What I find of particular interest is finding *non-kinship* terms that
exhibit structural ambiguity, rather than simple underdeterminacy.
Without them, this discussion merely boils down to semantics of kinship.
Comparative semantics of kinship makes for an interesting linguistic and
anthropological discussion in its own right, but non-kinship words
elevate it to another level. Consider this a challenge. For example, in
English, "ham" is generally just underdetermined (raw, boiled, baked,
smoked, etc.), but, as far as I can tell, it may be ambiguous in Dutch
(as to foreleg vs. hind leg). There should be easier and clearer examples.


Laurence Horn wrote:
> At 1:22 AM -0800 1/22/09, Benjamin Barrett wrote:
>> Brother and sister as well as uncle and aunt are ambiguous in some
>> cultures as you have to indicate younger or older. I think Cantonese,
>> for example, has a number of these. Japanese has the bizarre case
>> where "cousin" is pronounced as "itoko" but is written four different
>> ways depending on whether the cousin is older or younger, male or
>> female.
> I would submit that these aren't actual ambiguities, but instances of
> vagueness or underspecification.  The standard identity-of-sense
> tests for ambiguity ("I have three uncles" vs. "I visited two banks"
> or "Neither Sally nor Beth can bear children") don't respect these
> differences in ways that someone can be an uncle or sister or
> brother-in-law, and the fact that other languages make a distinction
> we don't isn't decisive.
>> ...
>> Another sort of ambiguity is "my grandmother," one that implies that
>> you have only one living, though that's not necessarily the case.
> Any more than "my sister"?  I can say I got an e-mail from my sister
> about X without implying that I have just one sister.  (Actually, I
> have none, but...)
> LH

The American Dialect Society -

More information about the Ads-l mailing list